Hi,
I have a tool that produces a variable number of outputs, which are written to the tmp database path with the specially formatted name "%s_%s_%s_%s_%s" % ( 'primary', output1.id, name, 'visible', file_type ). In lib/galaxy/tools/__init__.py, the history item's "designation" is set to "name" and the item's "name" is set to output1.name. This potentially creates many items with the same name, which must be differentiated either through meta data or by examining the contents of the file. I would like to request a means to customize the name of the collected dataset.
A simple but hack-y method would be to append another token to the name of the file. If this token exists, it would be used as the history item's name e.g. "primary_123_NAME123_visible_txt_dbkey_NEWNAME". This would also require that the optional DBKEY token would have to be validated for non-blank values.
Josh